Integration of viral into host cell DNA is a central and unique feature of retrovirus replication, and is directly responsible for many of the special characteristics of this important group of viruses, including their ability to cause cancer by activating and transducing oncogenes, their evolutionary persistence as endogenous viruses in the host germline, their very broad pathogenic spectrum, and their ability to serve as vectors for the gene therapist. Although the biochemical and structural aspects of the integration process are becoming quite well worked out, many important features remain unknown, including what defines an integration target, the relationship between chromatin and DNA structure and integration specificity, and the effect of the location of the integrated provirus on its subsequent expression. The overall goal of this project is to improve our understanding of key aspects of the role of integration in the virus-host interaction. Specifically, we will address the following questions: 1. Can we define an optimal target sequence (or structure) for integrase in vitro? What are the structural correlates of integrase targeting? How does the intracellular environment affect the use of certain sequences (or structures) as targets? 2. How does the structure, location, and activity of a region of cellular DNA affect its use as an integration target? In particular, we will test the hypothesis, based on our prior work, that integration has no strong regional preference, and that (contrary to prior ideas), increasing transcriptional activity of a gene does not enhance is use as an integration target, but rather reduces it, as a consequence of interference of bound transcription factors. 3. How does integration affect subsequent expression of the provirus. Why is integration apparently required for efficient expression? What is the role of the integration site in the level of transcription of the integrated provirus? In particular, what is the basis for "position effects" leading to clone-to-clone variation in expression of integrated proviruses?